LLM-CLI

What is LLM-CLI?

LLM-CLI is a command-line AI assistant that supports multiple large language models (LLMs). It is a fork of Gemini Cli but offers broader capabilities by supporting various providers like Gemini, OpenAI, Anthropic Claude, and Ollama. This tool is designed to be a powerful agent for coding and other tasks, allowing users to leverage their preferred LLM, whether it's running in the cloud or locally.

Benefits

LLM-CLI offers several key advantages:

• Multi-provider support: Use your preferred LLM provider, including OpenAI, Anthropic, Ollama, and Gemini.
• Interactive CLI interface: Enjoy a user-friendly command-line experience with full tool and function support.
• Context-aware file processing: Process files with context awareness for more accurate and relevant results.
• Extensible with custom commands: Customize the tool to fit your specific needs by adding custom commands.
• Provider-specific model switching: Easily switch between different LLM providers and models.
• Memory and context management: Maintain context and memory for more coherent and continuous interactions.
• Secure API key management: Manage your API keys securely with persistent settings.
• Full tool/function calling support: Utilize all available CLI tools, including file operations, shell commands, and web searches.

Use Cases

LLM-CLI is versatile and can be used in various scenarios:

• Coding assistance: Get help with coding tasks, from writing and debugging code to understanding complex concepts.
• File processing: Process and analyze files with context awareness for better accuracy.
• Automation: Automate repetitive tasks by integrating LLM-CLI with other tools and commands.
• Research and learning: Use the tool to gather information, summarize texts, and learn new topics.
• Custom workflows: Create custom workflows by extending LLM-CLI with custom commands tailored to your needs.

Installation

To install LLM-CLI, use the following command:

npm install -g @marizmelo/llm-cli

Usage

To start using LLM-CLI, simply type:

llm-cli

For non-interactive mode, you can use:

llm-cli --prompt "What is 2+2?"

What's New in v0.1.0

The latest version of LLM-CLI includes several enhancements:

• Enhanced provider support: Full function/tool calling capabilities for OpenAI and Anthropic, allowing the use of all available CLI tools.
• Persistent API key management: API keys are now saved to settings and automatically synced to environment variables on startup.
• Improved provider setup: Simplified setup process with the/provider setup <provider> <api-key>command.
• Better error handling: Fixed streaming response issues and improved error messages.

Provider Setup

Quick Setup (New Method)

# Setup providers with persistent API keysllm-cli/provider setup openai sk-your-api-key/provider setup anthropic sk-ant-your-api-key/provider switch openai

Ollama (Local)

# Install Ollama first, then:llm-cli/provider switch ollama

Traditional Setup (Environment Variables)

# Still supported for backward compatibilityexport OPENAI_API_KEY="your-api-key"export ANTHROPIC_API_KEY="your-api-key"llm-cli

Documentation

For full documentation, visit:LLM-CLI Documentation

License

LLM-CLI is licensed under Apache-2.0.

About

LLM-CLI is an open-source AI agent that brings the power of Cloud and Local LLMs directly into your terminal.